Introduction

Abstract

Experiments are performed in three general steps: first, the experiment must be designed; second, the data must be gathered; and third, the data must be analyzed. These three steps are highly idealized, and no clear boundary exists between them. The problem of analyzing the data is one that should be faced early in the design phase. Gathering the data in such a way as to learn the most about a phenomenon is what doing an experiment is all about. It will do an experimenter little good to obtain a set of data that does not bear directly on the model, or hypotheses, to be tested.